SHARE:

Media trust and infection mitigating behaviours during the COVID-19 pandemic in the USA

General Information

Title
Media trust and infection mitigating behaviours during the COVID-19 pandemic in the USA
Author
Erfei Zhao, Qiao Wu, Eileen M Crimmins and Jennifer A Ailshire
Publication Type
Journal paper
Outlet
BMJ Global Health
Year
2020
Abstract
Introduction

The COVID-19 pandemic is an unprecedented public health crisis. It is becoming increasingly clear that people’s behavioural responses in the USA during this fast-changing pandemic are associated with their preferred media sources. The polarisation of US media has been reflected in politically motivated messaging around the coronavirus by some media outlets, such as Fox News. This resulted in different messaging around the risks of infection and behavioural changes necessary to mitigate that risk. This study determined if COVID-related behaviours differed according to trust in left-leaning or right-leaning media and how differences changed over the first several months of the pandemic.

Methods

Using the nationally representative Understanding America Study COVID-19 panel, we examine preventive and risky behaviours related to infection from COVID-19 over the period from 10 March to 9 June for people with trust in different media sources: one left-leaning, CNN and another right-leaning, Fox News. People’s media preferences are categorised into three groups: (1) those who trust CNN more than Fox News; (2) those who have equal or no preferences and (3) those who trust Fox News more than CNN.

Results

Results showed that compared with those who trust CNN more than Fox news, people who trust Fox News more than CNN engaged in fewer preventive behaviours and more risky behaviours related to COVID-19. Out of five preventive and five risky behaviours examined, people who trust Fox News more than CNN practised an average of 3.41 preventive behaviours and 1.25 risky behaviours, while those who trust CNN more than Fox News engaged in an average of 3.85 preventive and 0.94 risky behaviours, from late March to June. The difference between these two groups widened in the month of May (p≤0.01), even after controlling for access to professional information and overall diversity of information sources.

Conclusions

Our findings indicate that behavioural responses were divided along media bias lines. In such a highly partisan environment, false information can be easily disseminated, and health messaging, which is one of the few effective ways to slowdown the spread of the virus in the absence of a vaccine, is being damaged by politically biased and economically focused narratives. During a public health crisis, media should reduce their partisan stance on health information, and the health messaging from neutral and professional sources based on scientific findings should be better promoted.